© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Attention SAE Research Paper
    5. GPT2-Small
    7. Attention Out
    9. 9-ATT-KK
    11. 22428
    Prev
    Next
    INDEX
    Explanations

    mentions of names or identifiers, particularly in a gaming or online context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-att-kk
    Configuration
    ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    float32
    Hook Name
    blocks.9.attn.hook_z
    Hook Layer
    9
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Head Attr Weights
    0:0.26
    1:0.02
    2:0.10
    3:0.08
    4:0.04
    5:0.07
    6:0.04
    7:0.05
    8:0.14
    9:0.03
    10:0.07
    11:0.04
    Negative Logits
     Rouhani
    -2.73
     Yemeni
    -2.60
     Tehran
    -2.51
     Shir
    -2.45
     projection
    -2.44
     Yamaha
    -2.43
    �
    -2.35
    511
    -2.30
     projections
    -2.29
     buds
    -2.28
    POSITIVE LOGITS
    roo
    3.64
    doms
    2.85
     McD
    2.75
     Duck
    2.66
     OCD
    2.46
    olean
    2.45
     hoard
    2.44
     Ng
    2.43
     divest
    2.38
     plantations
    2.33
    Activations Density 0.000%

    No Known Activations